MCC生成的I2C代码有一个错误

你确定吗？那之后直接发生了什么？第一个窗体（“不执行循环”）是一种常见的等待特定状态标志更改的方式。



      以下为原文

    Are you sure?
What comes straight after that?
The first form ( a "do nothing loop") is a common way to wait until a particular status flag changes.
 

我上传更多的代码用于理解。//用于中断的IF（*pFLAG=＝I2C1YMasaGug等待）{（I2C1LY状态）！{s/MistelyDelphi；{//强迫任务运行，因为我们知道队列有/ /需要发送PIR1BITS。SSP1IF＝TRUE；} //块，直到请求完成为止。



      以下为原文

    I upload more code for understanding.
 
    // for interrupt based
    if (*pflag == I2C1_MESSAGE_PENDING)
    {
        while(i2c1_state != S_MASTER_IDLE);
        {
            // force the task to run since we know that the queue has
            // something that needs to be sent
            PIR1bits.SSP1IF = true;
        }
    }   // block until request is complete

我上传更多的代码用于理解。//用于中断的IF（*pFLAG=＝I2C1YMasaGug等待）{（I2C1LY状态）！{s/MistelyDelphi；{//强迫任务运行，因为我们知道队列有/ /需要发送PIR1BITS。SSP1IF＝TRUE；} //块，直到请求完成为止。



      以下为原文

    I upload more code for understanding.
 
    // for interrupt based
    if (*pflag == I2C1_MESSAGE_PENDING)
    {
        while(i2c1_state != S_MASTER_IDLE);
        {
            // force the task to run since we know that the queue has
            // something that needs to be sent
            PIR1bits.SSP1IF = true;
        }
    }   // block until request is complete

完全正常。while行等待开始条件设置，然后下一个块设置SSPIF。它不尝试多次设置SSPIF，直到开始完成。



      以下为原文

    Perfectly normal.
The while line waits for the Start condition to be set, then the next block sets SSPIF.
 
It is NOT trying to set SSPIF multiple times until the Start completes.

然后我想这样写。//用于中断的IF（*pFLAG=＝I2C1YMasaGug等待）{（I2C1LY状态）！（= SyMistelIDLE）；/ /强制执行任务，因为我们知道队列有/需要发送PIR1BITS。SSP1IF＝TRUE；} /块，直到请求完成为止。



      以下为原文

    Then I would like to write like this.
 
    // for interrupt based
    if (*pflag == I2C1_MESSAGE_PENDING)
    {
        while(i2c1_state != S_MASTER_IDLE);
        // force the task to run since we know that the queue has
        // something that needs to be sent
       PIR1bits.SSP1IF = true;
    }   // block until request is complete
 

我发布了这个帖子的根本原因。请参阅HTTP://www. McCHIP.COM/FUMUS/M1001673.ASPXY感谢您，丹尼尔。K



      以下为原文

    I posted the root cause of this post. Please see http://www.microchip.com/forums/m1001673.aspx
Thank you,
Daniel. K

是的，我最近测试过MCC，很方便，但是质量代码很差，至少这个I2C主代码不是一个健壮的驱动程序（PIC18）：1。可能的空指针写入，如果上拉电阻器失败或其他，Pi2iC1YOLD可以是空指针，因为初始化是在这个代码之后。2。“静态”I2C1TyActhOngRealEclipse块TrBug；在FuncI2C1MrimeWrrad中。它仅适用于I2C1CONTIONTRAIQueLeSyLange= 1。所以为什么Trb阙？UE在这个驱动程序中使用吗？



      以下为原文

    yes, I've test MCC recently, convenient but seems bad quality codes, at least this I2C master code is not a robust driver(PIC18):
 
1. possible NULL pointer written ,  if pull up resistor failed or something else,
   p_i2c1_current could be NULL pointer, since initialization is after this codes.
   // Check first if there was a collision.
    // If we have a Write Collision, reset and go to idle state */
    if(I2C1_WRITE_COLLISION_STATUS_BIT)
    {
        // clear the Write colision
        I2C1_WRITE_COLLISION_STATUS_BIT = 0;
        i2c1_state = S_MASTER_IDLE;
        *(p_i2c1_current->pTrFlag) = I2C1_MESSAGE_FAIL;

        // reset the buffer pointer
        p_i2c1_current = NULL;

        return;
    }
 
2. 
"static" I2C1_TRANSACTION_REQUEST_BLOCK   trBlock; in func I2C1_MasterWrite.
It's valid only with I2C1_CONFIG_TR_QUEUE_LENGTH = 1.
so why TRB queue is used in this driver?